Gaussian Process Based Bayesian Inference System for Intelligent Surface Measurement

Ren, Ming; Cheung, Chi Fai; Xiao, Gao

doi:10.3390/s18114069

Cited by 5 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GPs are a mathematical tool used to describe normally distributed stochastic processes that evolve in time according to probabilistic laws [37]. Each GP is a collection of random variables, any finite subset of which obeys a joint Gaussian distribution [38]. A GP is defined with the expression:…”

Section: Gaussian Process Algorithmsmentioning

confidence: 99%

Applications of data fusion in optical coordinate metrology: a review

Zhang

Catalucci

Thompson

et al. 2022

Int J Adv Manuf Technol

View full text Add to dashboard Cite

Data fusion enables the characterisation of an object using multiple datasets collected by various sensors. To improve optical coordinate measurement using data fusion, researchers have proposed numerous algorithmic solutions and methods. The most popular examples are the Gaussian process (GP) and weighted least-squares (WLS) algorithms, which depend on user-defined mathematical models describing the geometric characteristics of the measured object. Existing research on GP and WLS algorithms indicates that GP algorithms have been widely applied in both academia and industry, despite their use being limited to applications on relatively simple geometries. Research on WLS algorithms is less common than research on GP algorithms, as the mathematical tools used in the WLS cases are too simple to be applied with complex geometries. Machine learning is a new technology that is increasingly being applied to data fusion applications. Research on this technology is relatively scarce, but recent work has highlighted the potential of machine learning methods with significant results. Unlike GP and WLS algorithms, machine learning algorithms can autonomously learn the geometrical features of an object. To understand existing research in-depth and explore a path for future work, a new taxonomy of data fusion algorithms is proposed, covering the mathematical background and existing research surrounding each algorithm type. To conclude, the advantages and limitations of the existing methods are reviewed, highlighting the issues related to data quality and the types of test artefacts.

show abstract

Section: Gaussian Process Algorithmsmentioning

confidence: 99%

Applications of data fusion in optical coordinate metrology: a review

Zhang

Catalucci

Thompson

et al. 2022

Int J Adv Manuf Technol

View full text Add to dashboard Cite

show abstract

“…Starting with the measurement aspects, they can be used to advance scanning methodology, for example adaptive scanning, like in Gwyscan library [ 8 ] focusing on sampling data for optimal collection of statistical information about roughness. Similarly, generated data can be used for development of even more advanced sampling techniques, e.g., based on compressed sensing [ 9 , 10 ]. In contrast to measured data, generated datasets allow estimating the impact of different error sources on the algorithm performance in a more systematic manner.…”

Section: Introductionmentioning

confidence: 99%

Synthetic Data in Quantitative Scanning Probe Microscopy

Nečas

Klapetek

2021

Nanomaterials

View full text Add to dashboard Cite

Synthetic data are of increasing importance in nanometrology. They can be used for development of data processing methods, analysis of uncertainties and estimation of various measurement artefacts. In this paper we review methods used for their generation and the applications of synthetic data in scanning probe microscopy, focusing on their principles, performance, and applicability. We illustrate the benefits of using synthetic data on different tasks related to development of better scanning approaches and related to estimation of reliability of data processing methods. We demonstrate how the synthetic data can be used to analyse systematic errors that are common to scanning probe microscopy methods, either related to the measurement principle or to the typical data processing paths.

show abstract

“…Coussement et al [14] proposed a Bayesian decision support framework that formalizes subjective human expert opinions with more objective organizational information. Ren et al [15] proposed a Bayesian inference system based on a Gaussian process for realizing intelligent surface measurements of multi-sensor instruments. Chen et al [16] proposed a hybrid expert system to simulate the decisionmaking process of clinical sleep staging through symbol fusion.…”

Section: Introductionmentioning

confidence: 99%

A Bearing Fault Diagnosis Method Based on Feature Selection Feedback Network and Improved D-S Evidence Fusion

Tang

Wang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Bearings running state affects the normal operation of mechanical equipment. It is of great theoretical and practical value to carry out bearing fault diagnosis. In bearing fault diagnosis research, the extraction and selection of fault features can help improving the accuracy of bearing fault diagnosis. However, these researches suffer from the following weaknesses. (1) High dimension of the selected features.(2) Uncertainty of single sensor for data sampling. Therefore, in this paper, a feature selection feedback network (FSFN) is proposed to overcome the first weakness. At the same time, we proposed an improved Dempster-Shafer (IDS) evidence theory fusion method based on the kappa coefficient to deal with the second weakness. Extensive evaluations of the proposed method on the CUT-2 experimental platform dataset showed that FSFN can not only reduce the dimension of the final selected feature without decreasing the diagnostic accuracy but also shorten the time of feature selection. Moreover, compared with the existing DS evidence theory fusion method, IDS can achieve higher average fusion precision and improve the accuracy and reliability of bearing fault diagnosis.

show abstract

Gaussian Process Based Bayesian Inference System for Intelligent Surface Measurement

Cited by 5 publications

References 32 publications

Applications of data fusion in optical coordinate metrology: a review

Applications of data fusion in optical coordinate metrology: a review

Synthetic Data in Quantitative Scanning Probe Microscopy

A Bearing Fault Diagnosis Method Based on Feature Selection Feedback Network and Improved D-S Evidence Fusion

Contact Info

Product

Resources

About